Monitoring adherence and abstinence of cannabis use disorder patients: profile identification and relationship with long term treatment outcomes

Abstract

Background: Patients with cannabis use disorder (CUD) show heterogeneous sociodemographic and consumption patterns. Although previous studies, focused on identifying subgroups of CUD patients using input variables, have yielded useful results for planning individualized treatments, no published research has analyzed the profiles of CUD patients according to their therapeutic progress. This study therefore aims to identify subgroups of patients using adherence and abstinence indicators and to explore whether these profiles are associated with sociodemographic characteristics, consumption variables, and long-term therapeutic outcomes. Methods: This was a retrospective observational study with a multisite sample of 2055 CUD outpatients who were beginning treatment. The study monitored patient data at two-year follow-up. We conducted latent profiles analysis on the appointment attendance ratio and percentage of negative cannabis tests. Results: A three profile solution emerged: i) moderate abstinence/moderate adherence (n = 997); ii) high abstinence/moderate adherence (n = 613); and iii) high abstinence/high adherence (n = 445). The study found the most marked differences at the beginning of treatment for education level (chi2 (8) = 121.70, p < .001), source of referral (chi2 (12) = 203.55, p < .001), and frequency of cannabis use (chi2 (10) = 232.39, p < .001). Eighty percent of patients from the “high abstinence/high adherence” group were relapse-free at two year follow- up. This percentage decreased to 24.3 % in the “moderate abstinence/moderate adherence” group. Conclusions: Research has shown adherence and abstinence indicators to be useful for identifying subgroups of patients with different prognoses regarding long-term success. Recognizing the sociodemographic and con- sumption variables associated with these profiles at the beginning of treatment could help to inform the design of more individualized interventions